Abstract:The rapid evolution of deep learning has led to significant achievements in computer vision, primarily driven by complex convolutional neural networks (CNNs). However, the increasing depth and parameter count of these networks often result in overfitting and elevated computational demands. Knowledge distillation (KD) has emerged as a promising technique to address these issues by transferring knowledge from a large, well-trained teacher model to a more compact student model. This paper introduces a novel knowledge distillation method that simplifies the distillation process and narrows the performance gap between teacher and student models without relying on intricate knowledge representations. Our approach leverages a unique teacher network architecture designed to enhance the efficiency and effectiveness of knowledge transfer. Additionally, we introduce a streamlined teacher network architecture that transfers knowledge effectively through a simplified distillation process, enabling the student model to achieve high accuracy with reduced computational demands. Comprehensive experiments conducted on the CIFAR-10 dataset demonstrate that our proposed model achieves superior performance compared to traditional KD methods and established architectures such as ResNet and VGG networks. The proposed method not only maintains high accuracy but also significantly reduces training and validation losses. Key findings highlight the optimal hyperparameter settings (temperature T = 15.0 and smoothing factor α = 0.7), which yield the highest validation accuracy and lowest loss values. This research contributes to the theoretical and practical advancements in knowledge distillation, providing a robust framework for future applications and research in neural network compression and optimization. The simplicity and efficiency of our approach pave the way for more accessible and scalable solutions in deep learning model deployment.

B-AT-KD: Binary attention map knowledge distillation

DCCD: Reducing Neural Network Redundancy Via Distillation

Research on Knowledge Distillation Algorithm of Object Detection

Class Attention Transfer Based Knowledge Distillation

Adaptive Cross-Architecture Mutual Knowledge Distillation

Knowledge Distillation with Feature Maps for Image Classification

Attention and feature transfer based knowledge distillation

Channel Distillation: Channel-Wise Attention for Knowledge Distillation

Hybrid mix-up contrastive knowledge distillation

Simplified Knowledge Distillation for Deep Neural Networks Bridging the Performance Gap with a Novel Teacher–Student Architecture

Graph-based Knowledge Distillation by Multi-head Attention Network

SAKD: Sparse attention knowledge distillation

LAKD-Activation Mapping Distillation Based on Local Learning

BD-KD: Balancing the Divergences for Online Knowledge Distillation

A-A KD: Attention and Activation Knowledge Distillation

Self-knowledge distillation enhanced binary neural networks derived from underutilized information

Channel-wise Knowledge Distillation for Dense Prediction

From Knowledge Distillation to Self-Knowledge Distillation: A Unified Approach with Normalized Loss and Customized Soft Labels

DE-MKD: Decoupled Multi-Teacher Knowledge Distillation Based on Entropy

Rethinking Knowledge Distillation Via Cross-Entropy

Multiple-Stage Knowledge Distillation